6dbc01e801
This is a NFC patch. This patch run a bulk update on CodeGen tests that are impacted by the true16 features. This patch applies: 1. duplicate GFX11plus runlines and apply them with "+mattr=+real-true16" and "+mattr=-real-true16" 2. update the test with the update script For some GISEL runlines, the current CodeGen do not fully support the true16 version. Still update the runlines, but comment out the failing one, and added a "FIXME-TRUE16" comment to that test for easier tracking. These test will be fixed in the following patches. This is in a transition state that we support both "+real-true16/-real-true16" in our code base. We plan to move to "+real-true16" as default, and finally remove "-real-true16" mode and test lines.
3366 lines
137 KiB
LLVM
3366 lines
137 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx803 < %s | FileCheck -check-prefix=GFX8 %s
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 < %s | FileCheck -check-prefix=GFX9 %s
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1030 < %s | FileCheck -check-prefix=GFX10 %s
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -mattr=+real-true16 < %s | FileCheck -check-prefixes=GFX11,GFX11-TRUE16 %s
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -mattr=-real-true16 < %s | FileCheck -check-prefixes=GFX11,GFX11-FAKE16 %s
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1200 -mattr=+real-true16 < %s | FileCheck -check-prefixes=GFX12,GFX12-TRUE16 %s
|
|
; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1200 -mattr=-real-true16 < %s | FileCheck -check-prefixes=GFX12,GFX12-FAKE16 %s
|
|
|
|
define half @v_maximumnum_f16(half %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.h, v1.l, v1.l
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.h, v1.l, v1.l
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v1, v1, v1
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call half @llvm.maximumnum.f16(half %x, half %y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_nnan(half %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v1.l
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v1.l
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_nnan:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan half @llvm.maximumnum.f16(half %x, half %y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_1.0(half %x) {
|
|
; GFX8-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f16_e32 v0, 1.0, v0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f16_e32 v0, 1.0, v0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, 1.0, v0.l
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, 1.0, v0
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, 1.0, v0.l
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_1.0:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, 1.0, v0
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call half @llvm.maximumnum.f16(half %x, half 1.0)
|
|
ret half %result
|
|
}
|
|
|
|
define bfloat @v_maximumnum_bf16(bfloat %x, bfloat %y) {
|
|
; GFX8-LABEL: v_maximumnum_bf16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX8-NEXT: v_cmp_u_f32_e32 vcc, v2, v2
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; GFX8-NEXT: v_cmp_u_f32_e32 vcc, v2, v2
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v1, v1, v0, vcc
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX8-NEXT: v_cmp_gt_f32_e32 vcc, v3, v2
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX8-NEXT: v_mul_f32_e32 v2, 1.0, v2
|
|
; GFX8-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX8-NEXT: v_add_u32_e32 v3, vcc, v3, v2
|
|
; GFX8-NEXT: s_movk_i32 s4, 0x7fff
|
|
; GFX8-NEXT: v_add_u32_e32 v3, vcc, s4, v3
|
|
; GFX8-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX8-NEXT: v_cmp_u_f32_e32 vcc, v2, v2
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v2, v3, v4, vcc
|
|
; GFX8-NEXT: v_lshrrev_b32_e32 v3, 16, v2
|
|
; GFX8-NEXT: v_cmp_eq_u16_e32 vcc, 0, v0
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc
|
|
; GFX8-NEXT: v_cmp_eq_u16_e32 vcc, 0, v1
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; GFX8-NEXT: v_and_b32_e32 v1, 0xffff0000, v2
|
|
; GFX8-NEXT: v_cmp_eq_f32_e32 vcc, 0, v1
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_bf16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX9-NEXT: v_cmp_u_f32_e32 vcc, v2, v2
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; GFX9-NEXT: v_cmp_u_f32_e32 vcc, v3, v3
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v1, v1, v0, vcc
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX9-NEXT: v_cmp_gt_f32_e32 vcc, v2, v3
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX9-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX9-NEXT: s_movk_i32 s4, 0x7fff
|
|
; GFX9-NEXT: v_add3_u32 v3, v3, v2, s4
|
|
; GFX9-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX9-NEXT: v_cmp_u_f32_e32 vcc, v2, v2
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v2, v3, v4, vcc
|
|
; GFX9-NEXT: v_lshrrev_b32_e32 v3, 16, v2
|
|
; GFX9-NEXT: v_cmp_eq_u16_e32 vcc, 0, v0
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc
|
|
; GFX9-NEXT: v_cmp_eq_u16_e32 vcc, 0, v1
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; GFX9-NEXT: v_and_b32_e32 v1, 0xffff0000, v2
|
|
; GFX9-NEXT: v_cmp_eq_f32_e32 vcc, 0, v1
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_bf16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX10-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX10-NEXT: v_cmp_u_f32_e32 vcc_lo, v3, v3
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v1, v1, v0, vcc_lo
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX10-NEXT: v_cmp_gt_f32_e32 vcc_lo, v2, v3
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc_lo
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX10-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX10-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX10-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX10-NEXT: v_add3_u32 v3, v3, v2, 0x7fff
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v2, v3, v4, vcc_lo
|
|
; GFX10-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0
|
|
; GFX10-NEXT: v_lshrrev_b32_e32 v3, 16, v2
|
|
; GFX10-NEXT: v_and_b32_e32 v2, 0xffff0000, v2
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc_lo
|
|
; GFX10-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX10-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v2
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc_lo
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_bf16:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX11-TRUE16-NEXT: v_mov_b16_e32 v0.h, v1.l
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v1, 16, v1
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX11-TRUE16-NEXT: v_cmp_u_f32_e64 s0, v1, v1
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v0.l, v0.h, vcc_lo
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v1.l, v0.h, v0.l, s0
|
|
; GFX11-TRUE16-NEXT: v_mov_b16_e32 v2.l, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_mov_b16_e32 v3.l, v1.l
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v3, 16, v3
|
|
; GFX11-TRUE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v2, v3
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v2.l, v1.l, v0.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX11-TRUE16-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_3)
|
|
; GFX11-TRUE16-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX11-TRUE16-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX11-TRUE16-NEXT: v_cmp_u_f32_e64 s0, v2, v2
|
|
; GFX11-TRUE16-NEXT: v_add3_u32 v3, v3, v2, 0x7fff
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b32_e64 v2, v3, v4, s0
|
|
; GFX11-TRUE16-NEXT: v_cmp_eq_u16_e64 s0, 0, v1.l
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.h, v0.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_and_b32_e32 v3, 0xffff0000, v2
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v0.l, v1.l, s0
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.h, v0.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_bf16:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2)
|
|
; GFX11-FAKE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v3, v3
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v1, v1, v0, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v2, v3
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX11-FAKE16-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX11-FAKE16-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX11-FAKE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_add3_u32 v3, v3, v2, 0x7fff
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v2, v3, v4, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-FAKE16-NEXT: v_lshrrev_b32_e32 v3, 16, v2
|
|
; GFX11-FAKE16-NEXT: v_and_b32_e32 v2, 0xffff0000, v2
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_4)
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v2
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2)
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc_lo
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_bf16:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX12-TRUE16-NEXT: v_mov_b16_e32 v0.h, v1.l
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v1, 16, v1
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX12-TRUE16-NEXT: v_cmp_u_f32_e64 s0, v1, v1
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v0.l, v0.h, vcc_lo
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xf1ff
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v1.l, v0.h, v0.l, s0
|
|
; GFX12-TRUE16-NEXT: v_mov_b16_e32 v2.l, v0.l
|
|
; GFX12-TRUE16-NEXT: v_mov_b16_e32 v3.l, v1.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v3, 16, v3
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_3) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v2, v3
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v2.l, v1.l, v0.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0.l
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f32_e32 v2, v2, v2
|
|
; GFX12-TRUE16-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX12-TRUE16-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX12-TRUE16-NEXT: v_cmp_u_f32_e64 s0, v2, v2
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_add3_u32 v3, v3, v2, 0x7fff
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xf1ff
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b32_e64 v2, v3, v4, s0
|
|
; GFX12-TRUE16-NEXT: v_cmp_eq_u16_e64 s0, 0, v1.l
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.h, v0.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: v_and_b32_e32 v3, 0xffff0000, v2
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xf1ff
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v0.l, v1.l, s0
|
|
; GFX12-TRUE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.h, v0.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_bf16:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_3) | instid1(VALU_DEP_2)
|
|
; GFX12-FAKE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_2) | instid1(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v3, v3
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v1, v1, v0, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v1
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v2, v3
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v2
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f32_e32 v2, v2, v2
|
|
; GFX12-FAKE16-NEXT: v_bfe_u32 v3, v2, 16, 1
|
|
; GFX12-FAKE16-NEXT: v_or_b32_e32 v4, 0x400000, v2
|
|
; GFX12-FAKE16-NEXT: v_cmp_u_f32_e32 vcc_lo, v2, v2
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_add3_u32 v3, v3, v2, 0x7fff
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v2, v3, v4, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_2) | instid1(VALU_DEP_2)
|
|
; GFX12-FAKE16-NEXT: v_lshrrev_b32_e32 v3, 16, v2
|
|
; GFX12-FAKE16-NEXT: v_and_b32_e32 v2, 0xffff0000, v2
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_2) | instid1(VALU_DEP_2)
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v2
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v0, v3, v0, vcc_lo
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call bfloat @llvm.maximumnum.bf16(bfloat %x, bfloat %y)
|
|
ret bfloat %result
|
|
}
|
|
|
|
define bfloat @v_maximumnum_bf16_nnan(bfloat %x, bfloat %y) {
|
|
; GFX8-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX8-NEXT: v_cmp_gt_f32_e32 vcc, v3, v2
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc
|
|
; GFX8-NEXT: v_cmp_eq_u16_e32 vcc, 0, v0
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; GFX8-NEXT: v_cmp_eq_u16_e32 vcc, 0, v1
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; GFX8-NEXT: v_lshlrev_b32_e32 v1, 16, v2
|
|
; GFX8-NEXT: v_cmp_eq_f32_e32 vcc, 0, v1
|
|
; GFX8-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX9-NEXT: v_cmp_gt_f32_e32 vcc, v3, v2
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc
|
|
; GFX9-NEXT: v_cmp_eq_u16_e32 vcc, 0, v0
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; GFX9-NEXT: v_cmp_eq_u16_e32 vcc, 0, v1
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; GFX9-NEXT: v_lshlrev_b32_e32 v1, 16, v2
|
|
; GFX9-NEXT: v_cmp_eq_f32_e32 vcc, 0, v1
|
|
; GFX9-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX10-NEXT: v_cmp_gt_f32_e32 vcc_lo, v3, v2
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc_lo
|
|
; GFX10-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0
|
|
; GFX10-NEXT: v_lshlrev_b32_e32 v3, 16, v2
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc_lo
|
|
; GFX10-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX10-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX10-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc_lo
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v3, v2
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v2.l, v1.l, v0.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0.l
|
|
; GFX11-TRUE16-NEXT: v_mov_b16_e32 v3.l, v2.l
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.l, v0.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-TRUE16-NEXT: v_lshlrev_b32_e32 v3, 16, v3
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v0.l, v1.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-TRUE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX11-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.l, v0.l, vcc_lo
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX11-FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2)
|
|
; GFX11-FAKE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v3, v2
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0
|
|
; GFX11-FAKE16-NEXT: v_dual_cndmask_b32 v0, v2, v0 :: v_dual_lshlrev_b32 v3, 16, v2
|
|
; GFX11-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX11-FAKE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2)
|
|
; GFX11-FAKE16-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc_lo
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_3) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v3, v2
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v2.l, v1.l, v0.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0.l
|
|
; GFX12-TRUE16-NEXT: v_mov_b16_e32 v3.l, v2.l
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.l, v0.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_3)
|
|
; GFX12-TRUE16-NEXT: v_lshlrev_b32_e32 v3, 16, v3
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v0.l, v1.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX12-TRUE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX12-TRUE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-TRUE16-NEXT: v_cndmask_b16 v0.l, v2.l, v0.l, vcc_lo
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_bf16_nnan:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v1
|
|
; GFX12-FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_4) | instid1(VALU_DEP_2)
|
|
; GFX12-FAKE16-NEXT: v_cmp_gt_f32_e32 vcc_lo, v3, v2
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v2, v1, v0, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v0
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: v_dual_cndmask_b32 v0, v2, v0 :: v_dual_lshlrev_b32 v3, 16, v2
|
|
; GFX12-FAKE16-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc_lo
|
|
; GFX12-FAKE16-NEXT: v_cmp_eq_f32_e32 vcc_lo, 0, v3
|
|
; GFX12-FAKE16-NEXT: s_wait_alu 0xfffd
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2)
|
|
; GFX12-FAKE16-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc_lo
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan bfloat @llvm.maximumnum.bf16(bfloat %x, bfloat %y)
|
|
ret bfloat %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v1, v1, v1 :: v_dual_max_f32 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v1, v1, v1 :: v_dual_max_num_f32 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_nnan(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan float @llvm.maximumnum.f32(float %x, float %y)
|
|
ret float %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64(double %x, double %y) {
|
|
; GFX8-LABEL: v_maximumnum_f64:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[2:3]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[2:3]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call double @llvm.maximumnum.f64(double %x, double %y)
|
|
ret double %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64_nnan(double %x, double %y) {
|
|
; GFX8-LABEL: v_maximumnum_f64_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[2:3]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan double @llvm.maximumnum.f64(double %x, double %y)
|
|
ret double %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_1.0(float %x) {
|
|
; GFX8-LABEL: v_maximumnum_f32_1.0:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_1.0:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, 1.0, v0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_1.0:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, 1.0, v0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_1.0:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, 1.0, v0
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_1.0:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, 1.0, v0
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call float @llvm.maximumnum.f32(float %x, float 1.0)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_rhs_not_snan(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_rhs_not_snan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_rhs_not_snan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_rhs_not_snan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_rhs_not_snan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v1, v1, v1 :: v_dual_max_f32 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_rhs_not_snan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v1, v1, v1 :: v_dual_max_num_f32 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%canon.y = call float @llvm.canonicalize.f32(float %y)
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %canon.y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_lhs_not_snan(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_lhs_not_snan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_lhs_not_snan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_lhs_not_snan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_lhs_not_snan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v0 :: v_dual_max_f32 v1, v1, v1
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_lhs_not_snan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v0 :: v_dual_max_num_f32 v1, v1, v1
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%canon.x = call float @llvm.canonicalize.f32(float %x)
|
|
%result = call float @llvm.maximumnum.f32(float %canon.x, float %y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_both_operands_not_snan(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_both_operands_not_snan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_both_operands_not_snan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_both_operands_not_snan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_both_operands_not_snan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v0 :: v_dual_max_f32 v1, v1, v1
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_both_operands_not_snan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v0 :: v_dual_max_num_f32 v1, v1, v1
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%canon.x = call float @llvm.canonicalize.f32(float %x)
|
|
%canon.y = call float @llvm.canonicalize.f32(float %y)
|
|
%result = call float @llvm.maximumnum.f32(float %canon.x, float %canon.y)
|
|
ret float %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64_1.0(double %x) {
|
|
; GFX8-LABEL: v_maximumnum_f64_1.0:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], 1.0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64_1.0:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], 1.0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64_1.0:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], 1.0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64_1.0:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], 1.0
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64_1.0:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], 1.0, v[0:1]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call double @llvm.maximumnum.f64(double %x, double 1.0)
|
|
ret double %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_s_v(half inreg %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, s0, s0
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.h, v0.l
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, s0, s0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, s0, s0
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.h, v0.l
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_s_v:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, s0, s0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v1, v0
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call half @llvm.maximumnum.f16(half %x, half %y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_v_s(half %x, half inreg %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, s0, s0
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, s0, s0
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, s0, s0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_v_s:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, s0, s0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call half @llvm.maximumnum.f16(half %x, half %y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_s_s(half inreg %x, half inreg %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e64 v0, s17, s17
|
|
; GFX8-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e64 v0, s17, s17
|
|
; GFX9-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e64 v0, s17, s17
|
|
; GFX10-NEXT: v_max_f16_e64 v1, s16, s16
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.l, s1, s1
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, s0, s0
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.h, v0.l
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v0, s1, s1
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, s0, s0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v1, v0
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.l, s1, s1
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, s0, s0
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.h, v0.l
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_s_s:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v0, s1, s1
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, s0, s0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v1, v0
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call half @llvm.maximumnum.f16(half %x, half %y)
|
|
ret half %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_s_v(float inreg %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_s_v:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_mul_f32_e64 v1, 1.0, s16
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_s_v:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e64 v1, s16, s16
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_s_v:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e64 v1, s16, s16
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_s_v:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX11-NEXT: v_max_f32_e64 v1, s0, s0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_s_v:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, s0, s0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v1, v0
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_v_s(float %x, float inreg %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_v_s:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e64 v1, 1.0, s16
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_v_s:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e64 v1, s16, s16
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_v_s:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e64 v1, s16, s16
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_v_s:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e64 v1, s0, s0
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_v_s:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, s0, s0
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_s_s(float inreg %x, float inreg %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_s_s:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e64 v0, 1.0, s17
|
|
; GFX8-NEXT: v_mul_f32_e64 v1, 1.0, s16
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_s_s:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e64 v0, s17, s17
|
|
; GFX9-NEXT: v_max_f32_e64 v1, s16, s16
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_s_s:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e64 v0, s17, s17
|
|
; GFX10-NEXT: v_max_f32_e64 v1, s16, s16
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_s_s:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e64 v0, s1, s1
|
|
; GFX11-NEXT: v_max_f32_e64 v1, s0, s0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v1, v0
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_s_s:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v0, s1, s1
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, s0, s0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v1, v0
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %y)
|
|
ret float %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64_s_v(double inreg %x, double %y) {
|
|
; GFX8-LABEL: v_maximumnum_f64_s_v:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64_s_v:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64_s_v:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64_s_v:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[2:3], s[0:1], s[0:1]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64_s_v:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e64 v[2:3], s[0:1], s[0:1]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[2:3], v[0:1]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call double @llvm.maximumnum.f64(double %x, double %y)
|
|
ret double %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64_v_s(double %x, double inreg %y) {
|
|
; GFX8-LABEL: v_maximumnum_f64_v_s:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64_v_s:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64_v_s:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64_v_s:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[2:3], s[0:1], s[0:1]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64_v_s:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e64 v[2:3], s[0:1], s[0:1]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[2:3]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call double @llvm.maximumnum.f64(double %x, double %y)
|
|
ret double %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64_s_s(double inreg %x, double inreg %y) {
|
|
; GFX8-LABEL: v_maximumnum_f64_s_s:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], s[18:19], s[18:19]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64_s_s:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], s[18:19], s[18:19]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64_s_s:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[0:1], s[18:19], s[18:19]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], s[16:17], s[16:17]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64_s_s:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], s[2:3], s[2:3]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], s[0:1], s[0:1]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64_s_s:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e64 v[0:1], s[2:3], s[2:3]
|
|
; GFX12-NEXT: v_max_num_f64_e64 v[2:3], s[0:1], s[0:1]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[2:3], v[0:1]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call double @llvm.maximumnum.f64(double %x, double %y)
|
|
ret double %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_fabs_rhs(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_fabs_rhs:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e64 v1, 1.0, |v1|
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_fabs_rhs:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e64 v1, |v1|, |v1|
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_fabs_rhs:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e64 v1, |v1|, |v1|
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_fabs_rhs:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e64 v1, |v1|, |v1|
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_fabs_rhs:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, |v1|, |v1|
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%fabs.y = call float @llvm.fabs.f32(float %y)
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %fabs.y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_fneg_fabs_rhs(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_fneg_fabs_rhs:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e64 v1, -1.0, |v1|
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_fneg_fabs_rhs:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e64 v1, -|v1|, -|v1|
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_fneg_fabs_rhs:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e64 v1, -|v1|, -|v1|
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_fneg_fabs_rhs:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e64 v1, -|v1|, -|v1|
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_fneg_fabs_rhs:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, -|v1|, -|v1|
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%fabs.y = call float @llvm.fabs.f32(float %y)
|
|
%fneg.fabs.y = fneg float %fabs.y
|
|
%result = call float @llvm.maximumnum.f32(float %x, float %fneg.fabs.y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_fabs(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_fabs:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e64 v1, 1.0, |v1|
|
|
; GFX8-NEXT: v_mul_f32_e64 v0, 1.0, |v0|
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_fabs:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e64 v1, |v1|, |v1|
|
|
; GFX9-NEXT: v_max_f32_e64 v0, |v0|, |v0|
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_fabs:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e64 v1, |v1|, |v1|
|
|
; GFX10-NEXT: v_max_f32_e64 v0, |v0|, |v0|
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_fabs:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e64 v1, |v1|, |v1|
|
|
; GFX11-NEXT: v_max_f32_e64 v0, |v0|, |v0|
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_fabs:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, |v1|, |v1|
|
|
; GFX12-NEXT: v_max_num_f32_e64 v0, |v0|, |v0|
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%fabs.x = call float @llvm.fabs.f32(float %x)
|
|
%fabs.y = call float @llvm.fabs.f32(float %y)
|
|
%result = call float @llvm.maximumnum.f32(float %fabs.x, float %fabs.y)
|
|
ret float %result
|
|
}
|
|
|
|
define float @v_maximumnum_f32_fneg(float %x, float %y) {
|
|
; GFX8-LABEL: v_maximumnum_f32_fneg:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, -1.0, v1
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, -1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f32_fneg:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e64 v1, -v1, -v1
|
|
; GFX9-NEXT: v_max_f32_e64 v0, -v0, -v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f32_fneg:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e64 v1, -v1, -v1
|
|
; GFX10-NEXT: v_max_f32_e64 v0, -v0, -v0
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f32_fneg:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f32_e64 v1, -v1, -v1
|
|
; GFX11-NEXT: v_max_f32_e64 v0, -v0, -v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f32_e32 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f32_fneg:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f32_e64 v1, -v1, -v1
|
|
; GFX12-NEXT: v_max_num_f32_e64 v0, -v0, -v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f32_e32 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%fneg.x = fneg float %x
|
|
%fneg.y = fneg float %y
|
|
%result = call float @llvm.maximumnum.f32(float %fneg.x, float %fneg.y)
|
|
ret float %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_fabs_rhs(half %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, |v1.l|, |v1.l|
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, |v1.l|, |v1.l|
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_fabs_rhs:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, |v1|, |v1|
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%fabs.y = call half @llvm.fabs.f16(half %y)
|
|
%result = call half @llvm.maximumnum.f16(half %x, half %fabs.y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_fneg_fabs_rhs(half %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e64 v1, -|v1|, -|v1|
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e64 v1, -|v1|, -|v1|
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e64 v1, -|v1|, -|v1|
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, -|v1.l|, -|v1.l|
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, -|v1|, -|v1|
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, -|v1.l|, -|v1.l|
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_fneg_fabs_rhs:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, -|v1|, -|v1|
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%fabs.y = call half @llvm.fabs.f16(half %y)
|
|
%fneg.fabs.y = fneg half %fabs.y
|
|
%result = call half @llvm.maximumnum.f16(half %x, half %fneg.fabs.y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_fabs(half %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX8-NEXT: v_max_f16_e64 v0, |v0|, |v0|
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX9-NEXT: v_max_f16_e64 v0, |v0|, |v0|
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX10-NEXT: v_max_f16_e64 v0, |v0|, |v0|
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, |v1.l|, |v1.l|
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.l, |v0.l|, |v0.l|
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, |v1|, |v1|
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v0, |v0|, |v0|
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, |v1.l|, |v1.l|
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.l, |v0.l|, |v0.l|
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_fabs:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, |v1|, |v1|
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v0, |v0|, |v0|
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%fabs.x = call half @llvm.fabs.f16(half %x)
|
|
%fabs.y = call half @llvm.fabs.f16(half %y)
|
|
%result = call half @llvm.maximumnum.f16(half %fabs.x, half %fabs.y)
|
|
ret half %result
|
|
}
|
|
|
|
define half @v_maximumnum_f16_fneg(half %x, half %y) {
|
|
; GFX8-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_e64 v1, -v1, -v1
|
|
; GFX8-NEXT: v_max_f16_e64 v0, -v0, -v0
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f16_e64 v1, -v1, -v1
|
|
; GFX9-NEXT: v_max_f16_e64 v0, -v0, -v0
|
|
; GFX9-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f16_e64 v1, -v1, -v1
|
|
; GFX10-NEXT: v_max_f16_e64 v0, -v0, -v0
|
|
; GFX10-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-TRUE16-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX11-TRUE16: ; %bb.0:
|
|
; GFX11-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.h, -v1.l, -v1.l
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e64 v0.l, -v0.l, -v0.l
|
|
; GFX11-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-TRUE16-NEXT: v_max_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX11-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-FAKE16-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX11-FAKE16: ; %bb.0:
|
|
; GFX11-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v1, -v1, -v1
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e64 v0, -v0, -v0
|
|
; GFX11-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-FAKE16-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX11-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-TRUE16-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX12-TRUE16: ; %bb.0:
|
|
; GFX12-TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-TRUE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.h, -v1.l, -v1.l
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e64 v0.l, -v0.l, -v0.l
|
|
; GFX12-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-TRUE16-NEXT: v_max_num_f16_e32 v0.l, v0.l, v0.h
|
|
; GFX12-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-FAKE16-LABEL: v_maximumnum_f16_fneg:
|
|
; GFX12-FAKE16: ; %bb.0:
|
|
; GFX12-FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-FAKE16-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v1, -v1, -v1
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e64 v0, -v0, -v0
|
|
; GFX12-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-FAKE16-NEXT: v_max_num_f16_e32 v0, v0, v1
|
|
; GFX12-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
%fneg.x = fneg half %x
|
|
%fneg.y = fneg half %y
|
|
%result = call half @llvm.maximumnum.f16(half %fneg.x, half %fneg.y)
|
|
ret half %result
|
|
}
|
|
|
|
define double @v_maximumnum_f64_fneg(double %x, double %y) {
|
|
; GFX8-LABEL: v_maximumnum_f64_fneg:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[2:3], -v[2:3], -v[2:3]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], -v[0:1], -v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_f64_fneg:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[2:3], -v[2:3], -v[2:3]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], -v[0:1], -v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_f64_fneg:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[2:3], -v[2:3], -v[2:3]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], -v[0:1], -v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_f64_fneg:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[2:3], -v[2:3], -v[2:3]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], -v[0:1], -v[0:1]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[2:3]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_f64_fneg:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e64 v[2:3], -v[2:3], -v[2:3]
|
|
; GFX12-NEXT: v_max_num_f64_e64 v[0:1], -v[0:1], -v[0:1]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[2:3]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%fneg.x = fneg double %x
|
|
%fneg.y = fneg double %y
|
|
%result = call double @llvm.maximumnum.f64(double %fneg.x, double %fneg.y)
|
|
ret double %result
|
|
}
|
|
|
|
define <2 x half> @v_maximumnum_v2f16(<2 x half> %x, <2 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v2f16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v2, v1, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v3, v0, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_sdwa v2, v3, v2 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v2
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v2f16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v2f16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v2f16:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v2f16:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v1
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v0
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <2 x half> @llvm.maximumnum.v2f16(<2 x half> %x, <2 x half> %y)
|
|
ret <2 x half> %result
|
|
}
|
|
|
|
define <2 x half> @v_maximumnum_v2f16_nnan(<2 x half> %x, <2 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v2f16_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v2, v0, v1 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v1
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v2
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v2f16_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v1
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v2f16_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v1
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v2f16_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v1
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v2f16_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v1
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <2 x half> @llvm.maximumnum.v2f16(<2 x half> %x, <2 x half> %y)
|
|
ret <2 x half> %result
|
|
}
|
|
|
|
define <3 x half> @v_maximumnum_v3f16(<3 x half> %x, <3 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v3f16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v4, v2, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v5, v0, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v2, v2, v2
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_sdwa v4, v5, v4 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_e32 v3, v3, v3
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v2
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v3
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v4
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v3f16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v3, v3
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v2
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v3f16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX10-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v3f16:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX11-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v3f16:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v2, v2, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v3, v3, v3
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v1
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v3
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <3 x half> @llvm.maximumnum.v3f16(<3 x half> %x, <3 x half> %y)
|
|
ret <3 x half> %result
|
|
}
|
|
|
|
define <3 x half> @v_maximumnum_v3f16_nnan(<3 x half> %x, <3 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v3f16_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v4, v0, v2 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v2
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v3
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v4
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v3f16_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v3f16_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v3f16_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v3f16_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v3
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <3 x half> @llvm.maximumnum.v3f16(<3 x half> %x, <3 x half> %y)
|
|
ret <3 x half> %result
|
|
}
|
|
|
|
define <4 x half> @v_maximumnum_v4f16(<4 x half> %x, <4 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v4f16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v4, v3, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v5, v1, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v4, v5, v4 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_sdwa v5, v2, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v6, v0, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v3, v3, v3
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX8-NEXT: v_max_f16_e32 v2, v2, v2
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_sdwa v5, v6, v5 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v3
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v2
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v5
|
|
; GFX8-NEXT: v_or_b32_e32 v1, v1, v4
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v4f16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v3, v3
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v2
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v4f16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX10-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v4f16:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX11-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v4f16:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v2, v2, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v3, v3, v3
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v1
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v3
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <4 x half> @llvm.maximumnum.v4f16(<4 x half> %x, <4 x half> %y)
|
|
ret <4 x half> %result
|
|
}
|
|
|
|
define <4 x half> @v_maximumnum_v4f16_nnan(<4 x half> %x, <4 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v4f16_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v4, v1, v3 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v5, v0, v2 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v3
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v2
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v5
|
|
; GFX8-NEXT: v_or_b32_e32 v1, v1, v4
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v4f16_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v4f16_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v4f16_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v4f16_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v3
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <4 x half> @llvm.maximumnum.v4f16(<4 x half> %x, <4 x half> %y)
|
|
ret <4 x half> %result
|
|
}
|
|
|
|
define <6 x half> @v_maximumnum_v6f16(<6 x half> %x, <6 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v6f16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v6, v5, v5 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v7, v2, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v6, v7, v6 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_sdwa v7, v4, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v8, v1, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v7, v8, v7 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_sdwa v8, v3, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v9, v0, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v5, v5, v5
|
|
; GFX8-NEXT: v_max_f16_e32 v2, v2, v2
|
|
; GFX8-NEXT: v_max_f16_e32 v4, v4, v4
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX8-NEXT: v_max_f16_e32 v3, v3, v3
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_sdwa v8, v9, v8 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_e32 v2, v2, v5
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v4
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v3
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v8
|
|
; GFX8-NEXT: v_or_b32_e32 v1, v1, v7
|
|
; GFX8-NEXT: v_or_b32_e32 v2, v2, v6
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v6f16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v3
|
|
; GFX9-NEXT: v_pk_max_f16 v3, v4, v4
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v3
|
|
; GFX9-NEXT: v_pk_max_f16 v3, v5, v5
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v2, v3
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v6f16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX10-NEXT: v_pk_max_f16 v4, v4, v4
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX10-NEXT: v_pk_max_f16 v5, v5, v5
|
|
; GFX10-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v3
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v4
|
|
; GFX10-NEXT: v_pk_max_f16 v2, v2, v5
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v6f16:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX11-NEXT: v_pk_max_f16 v4, v4, v4
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX11-NEXT: v_pk_max_f16 v5, v5, v5
|
|
; GFX11-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v3
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v4
|
|
; GFX11-NEXT: v_pk_max_f16 v2, v2, v5
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v6f16:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v3, v3, v3
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v4, v4, v4
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v1
|
|
; GFX12-NEXT: v_pk_max_num_f16 v5, v5, v5
|
|
; GFX12-NEXT: v_pk_max_num_f16 v2, v2, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v3
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v4
|
|
; GFX12-NEXT: v_pk_max_num_f16 v2, v2, v5
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <6 x half> @llvm.maximumnum.v6f16(<6 x half> %x, <6 x half> %y)
|
|
ret <6 x half> %result
|
|
}
|
|
|
|
define <8 x half> @v_maximumnum_v8f16(<8 x half> %x, <8 x half> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v8f16:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f16_sdwa v8, v7, v7 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v9, v3, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v8, v9, v8 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_sdwa v9, v6, v6 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v10, v2, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v9, v10, v9 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_sdwa v10, v5, v5 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v11, v1, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v10, v11, v10 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_sdwa v11, v4, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_sdwa v12, v0, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
|
|
; GFX8-NEXT: v_max_f16_e32 v7, v7, v7
|
|
; GFX8-NEXT: v_max_f16_e32 v3, v3, v3
|
|
; GFX8-NEXT: v_max_f16_e32 v6, v6, v6
|
|
; GFX8-NEXT: v_max_f16_e32 v2, v2, v2
|
|
; GFX8-NEXT: v_max_f16_e32 v5, v5, v5
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v1
|
|
; GFX8-NEXT: v_max_f16_e32 v4, v4, v4
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v0
|
|
; GFX8-NEXT: v_max_f16_sdwa v11, v12, v11 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
|
|
; GFX8-NEXT: v_max_f16_e32 v3, v3, v7
|
|
; GFX8-NEXT: v_max_f16_e32 v2, v2, v6
|
|
; GFX8-NEXT: v_max_f16_e32 v1, v1, v5
|
|
; GFX8-NEXT: v_max_f16_e32 v0, v0, v4
|
|
; GFX8-NEXT: v_or_b32_e32 v0, v0, v11
|
|
; GFX8-NEXT: v_or_b32_e32 v1, v1, v10
|
|
; GFX8-NEXT: v_or_b32_e32 v2, v2, v9
|
|
; GFX8-NEXT: v_or_b32_e32 v3, v3, v8
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v8f16:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_pk_max_f16 v4, v4, v4
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX9-NEXT: v_pk_max_f16 v0, v0, v4
|
|
; GFX9-NEXT: v_pk_max_f16 v4, v5, v5
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX9-NEXT: v_pk_max_f16 v1, v1, v4
|
|
; GFX9-NEXT: v_pk_max_f16 v4, v6, v6
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX9-NEXT: v_pk_max_f16 v2, v2, v4
|
|
; GFX9-NEXT: v_pk_max_f16 v4, v7, v7
|
|
; GFX9-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX9-NEXT: v_pk_max_f16 v3, v3, v4
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v8f16:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_pk_max_f16 v4, v4, v4
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX10-NEXT: v_pk_max_f16 v5, v5, v5
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX10-NEXT: v_pk_max_f16 v6, v6, v6
|
|
; GFX10-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX10-NEXT: v_pk_max_f16 v7, v7, v7
|
|
; GFX10-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX10-NEXT: v_pk_max_f16 v0, v0, v4
|
|
; GFX10-NEXT: v_pk_max_f16 v1, v1, v5
|
|
; GFX10-NEXT: v_pk_max_f16 v2, v2, v6
|
|
; GFX10-NEXT: v_pk_max_f16 v3, v3, v7
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v8f16:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_pk_max_f16 v4, v4, v4
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v0
|
|
; GFX11-NEXT: v_pk_max_f16 v5, v5, v5
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v1
|
|
; GFX11-NEXT: v_pk_max_f16 v6, v6, v6
|
|
; GFX11-NEXT: v_pk_max_f16 v2, v2, v2
|
|
; GFX11-NEXT: v_pk_max_f16 v7, v7, v7
|
|
; GFX11-NEXT: v_pk_max_f16 v3, v3, v3
|
|
; GFX11-NEXT: v_pk_max_f16 v0, v0, v4
|
|
; GFX11-NEXT: v_pk_max_f16 v1, v1, v5
|
|
; GFX11-NEXT: v_pk_max_f16 v2, v2, v6
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4)
|
|
; GFX11-NEXT: v_pk_max_f16 v3, v3, v7
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v8f16:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v4, v4, v4
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v0
|
|
; GFX12-NEXT: v_pk_max_num_f16 v5, v5, v5
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v1
|
|
; GFX12-NEXT: v_pk_max_num_f16 v6, v6, v6
|
|
; GFX12-NEXT: v_pk_max_num_f16 v2, v2, v2
|
|
; GFX12-NEXT: v_pk_max_num_f16 v7, v7, v7
|
|
; GFX12-NEXT: v_pk_max_num_f16 v3, v3, v3
|
|
; GFX12-NEXT: v_pk_max_num_f16 v0, v0, v4
|
|
; GFX12-NEXT: v_pk_max_num_f16 v1, v1, v5
|
|
; GFX12-NEXT: v_pk_max_num_f16 v2, v2, v6
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_4)
|
|
; GFX12-NEXT: v_pk_max_num_f16 v3, v3, v7
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <8 x half> @llvm.maximumnum.v8f16(<8 x half> %x, <8 x half> %y)
|
|
ret <8 x half> %result
|
|
}
|
|
|
|
define <2 x float> @v_maximumnum_v2f32(<2 x float> %x, <2 x float> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v2f32:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v2, 1.0, v2
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v2
|
|
; GFX8-NEXT: v_mul_f32_e32 v2, 1.0, v3
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_max_f32_e32 v1, v1, v2
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v2f32:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v2
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v3, v3
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v2
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v2f32:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v3, v3, v3
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v2
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v3
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v2f32:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v2, v2, v2 :: v_dual_max_f32 v3, v3, v3
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v0 :: v_dual_max_f32 v1, v1, v1
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v2 :: v_dual_max_f32 v1, v1, v3
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v2f32:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v2, v2, v2 :: v_dual_max_num_f32 v3, v3, v3
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v0 :: v_dual_max_num_f32 v1, v1, v1
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v2 :: v_dual_max_num_f32 v1, v1, v3
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <2 x float> @llvm.maximumnum.v2f32(<2 x float> %x, <2 x float> %y)
|
|
ret <2 x float> %result
|
|
}
|
|
|
|
define <2 x float> @v_maximumnum_v2f32_nnan(<2 x float> %x, <2 x float> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v2f32_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v2
|
|
; GFX8-NEXT: v_max_f32_e32 v1, v1, v3
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v2f32_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v2
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v3
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v2f32_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v2
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v3
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v2f32_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v2 :: v_dual_max_f32 v1, v1, v3
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v2f32_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v2 :: v_dual_max_num_f32 v1, v1, v3
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <2 x float> @llvm.maximumnum.v2f32(<2 x float> %x, <2 x float> %y)
|
|
ret <2 x float> %result
|
|
}
|
|
|
|
define <3 x float> @v_maximumnum_v3f32(<3 x float> %x, <3 x float> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v3f32:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v3, 1.0, v3
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v3
|
|
; GFX8-NEXT: v_mul_f32_e32 v3, 1.0, v4
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_max_f32_e32 v1, v1, v3
|
|
; GFX8-NEXT: v_mul_f32_e32 v3, 1.0, v5
|
|
; GFX8-NEXT: v_mul_f32_e32 v2, 1.0, v2
|
|
; GFX8-NEXT: v_max_f32_e32 v2, v2, v3
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v3f32:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v3, v3, v3
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v3
|
|
; GFX9-NEXT: v_max_f32_e32 v3, v4, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v3
|
|
; GFX9-NEXT: v_max_f32_e32 v3, v5, v5
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v3
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v3f32:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v3, v3, v3
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v4, v4, v4
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v5, v5, v5
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v3
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v4
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v5
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v3f32:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v3, v3, v3 :: v_dual_max_f32 v0, v0, v0
|
|
; GFX11-NEXT: v_dual_max_f32 v4, v4, v4 :: v_dual_max_f32 v1, v1, v1
|
|
; GFX11-NEXT: v_dual_max_f32 v5, v5, v5 :: v_dual_max_f32 v2, v2, v2
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v3 :: v_dual_max_f32 v1, v1, v4
|
|
; GFX11-NEXT: v_max_f32_e32 v2, v2, v5
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v3f32:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v3, v3, v3 :: v_dual_max_num_f32 v0, v0, v0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v4, v4, v4 :: v_dual_max_num_f32 v1, v1, v1
|
|
; GFX12-NEXT: v_dual_max_num_f32 v5, v5, v5 :: v_dual_max_num_f32 v2, v2, v2
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v3 :: v_dual_max_num_f32 v1, v1, v4
|
|
; GFX12-NEXT: v_max_num_f32_e32 v2, v2, v5
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <3 x float> @llvm.maximumnum.v3f32(<3 x float> %x, <3 x float> %y)
|
|
ret <3 x float> %result
|
|
}
|
|
|
|
define <3 x float> @v_maximumnum_v3f32_nnan(<3 x float> %x, <3 x float> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v3f32_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v3
|
|
; GFX8-NEXT: v_max_f32_e32 v1, v1, v4
|
|
; GFX8-NEXT: v_max_f32_e32 v2, v2, v5
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v3f32_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v3
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v5
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v3f32_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v3
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v4
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v5
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v3f32_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v3 :: v_dual_max_f32 v1, v1, v4
|
|
; GFX11-NEXT: v_max_f32_e32 v2, v2, v5
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v3f32_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v3 :: v_dual_max_num_f32 v1, v1, v4
|
|
; GFX12-NEXT: v_max_num_f32_e32 v2, v2, v5
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <3 x float> @llvm.maximumnum.v3f32(<3 x float> %x, <3 x float> %y)
|
|
ret <3 x float> %result
|
|
}
|
|
|
|
define <4 x float> @v_maximumnum_v4f32(<4 x float> %x, <4 x float> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v4f32:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_mul_f32_e32 v4, 1.0, v4
|
|
; GFX8-NEXT: v_mul_f32_e32 v0, 1.0, v0
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v4
|
|
; GFX8-NEXT: v_mul_f32_e32 v4, 1.0, v5
|
|
; GFX8-NEXT: v_mul_f32_e32 v1, 1.0, v1
|
|
; GFX8-NEXT: v_max_f32_e32 v1, v1, v4
|
|
; GFX8-NEXT: v_mul_f32_e32 v4, 1.0, v6
|
|
; GFX8-NEXT: v_mul_f32_e32 v2, 1.0, v2
|
|
; GFX8-NEXT: v_max_f32_e32 v2, v2, v4
|
|
; GFX8-NEXT: v_mul_f32_e32 v4, 1.0, v7
|
|
; GFX8-NEXT: v_mul_f32_e32 v3, 1.0, v3
|
|
; GFX8-NEXT: v_max_f32_e32 v3, v3, v4
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v4f32:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v4, v4, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v4, v5, v5
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v4, v6, v6
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v4, v7, v7
|
|
; GFX9-NEXT: v_max_f32_e32 v3, v3, v3
|
|
; GFX9-NEXT: v_max_f32_e32 v3, v3, v4
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v4f32:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v4, v4, v4
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v0
|
|
; GFX10-NEXT: v_max_f32_e32 v5, v5, v5
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v1
|
|
; GFX10-NEXT: v_max_f32_e32 v6, v6, v6
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v2
|
|
; GFX10-NEXT: v_max_f32_e32 v7, v7, v7
|
|
; GFX10-NEXT: v_max_f32_e32 v3, v3, v3
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v4
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v5
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v6
|
|
; GFX10-NEXT: v_max_f32_e32 v3, v3, v7
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v4f32:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v4, v4, v4 :: v_dual_max_f32 v5, v5, v5
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v0 :: v_dual_max_f32 v1, v1, v1
|
|
; GFX11-NEXT: v_dual_max_f32 v6, v6, v6 :: v_dual_max_f32 v7, v7, v7
|
|
; GFX11-NEXT: v_dual_max_f32 v2, v2, v2 :: v_dual_max_f32 v3, v3, v3
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v4 :: v_dual_max_f32 v1, v1, v5
|
|
; GFX11-NEXT: v_dual_max_f32 v2, v2, v6 :: v_dual_max_f32 v3, v3, v7
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v4f32:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v4, v4, v4 :: v_dual_max_num_f32 v5, v5, v5
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v0 :: v_dual_max_num_f32 v1, v1, v1
|
|
; GFX12-NEXT: v_dual_max_num_f32 v6, v6, v6 :: v_dual_max_num_f32 v7, v7, v7
|
|
; GFX12-NEXT: v_dual_max_num_f32 v2, v2, v2 :: v_dual_max_num_f32 v3, v3, v3
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v4 :: v_dual_max_num_f32 v1, v1, v5
|
|
; GFX12-NEXT: v_dual_max_num_f32 v2, v2, v6 :: v_dual_max_num_f32 v3, v3, v7
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <4 x float> @llvm.maximumnum.v4f32(<4 x float> %x, <4 x float> %y)
|
|
ret <4 x float> %result
|
|
}
|
|
|
|
define <4 x float> @v_maximumnum_v4f32_nnan(<4 x float> %x, <4 x float> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v4f32_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f32_e32 v0, v0, v4
|
|
; GFX8-NEXT: v_max_f32_e32 v1, v1, v5
|
|
; GFX8-NEXT: v_max_f32_e32 v2, v2, v6
|
|
; GFX8-NEXT: v_max_f32_e32 v3, v3, v7
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v4f32_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f32_e32 v0, v0, v4
|
|
; GFX9-NEXT: v_max_f32_e32 v1, v1, v5
|
|
; GFX9-NEXT: v_max_f32_e32 v2, v2, v6
|
|
; GFX9-NEXT: v_max_f32_e32 v3, v3, v7
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v4f32_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f32_e32 v0, v0, v4
|
|
; GFX10-NEXT: v_max_f32_e32 v1, v1, v5
|
|
; GFX10-NEXT: v_max_f32_e32 v2, v2, v6
|
|
; GFX10-NEXT: v_max_f32_e32 v3, v3, v7
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v4f32_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_dual_max_f32 v0, v0, v4 :: v_dual_max_f32 v1, v1, v5
|
|
; GFX11-NEXT: v_dual_max_f32 v2, v2, v6 :: v_dual_max_f32 v3, v3, v7
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v4f32_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_dual_max_num_f32 v0, v0, v4 :: v_dual_max_num_f32 v1, v1, v5
|
|
; GFX12-NEXT: v_dual_max_num_f32 v2, v2, v6 :: v_dual_max_num_f32 v3, v3, v7
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <4 x float> @llvm.maximumnum.v4f32(<4 x float> %x, <4 x float> %y)
|
|
ret <4 x float> %result
|
|
}
|
|
|
|
define <2 x double> @v_maximumnum_v2f64(<2 x double> %x, <2 x double> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v2f64:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v2f64:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v2f64:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v2f64:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v2f64:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[4:5]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[6:7], v[6:7], v[6:7]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[2:3]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[4:5]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[6:7]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <2 x double> @llvm.maximumnum.v2f64(<2 x double> %x, <2 x double> %y)
|
|
ret <2 x double> %result
|
|
}
|
|
|
|
define <2 x double> @v_maximumnum_v2f64_nnan(<2 x double> %x, <2 x double> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v2f64_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v2f64_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v2f64_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v2f64_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[6:7]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v2f64_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[4:5]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[6:7]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <2 x double> @llvm.maximumnum.v2f64(<2 x double> %x, <2 x double> %y)
|
|
ret <2 x double> %result
|
|
}
|
|
|
|
define <3 x double> @v_maximumnum_v3f64(<3 x double> %x, <3 x double> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v3f64:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX8-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v3f64:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX9-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v3f64:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX10-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v3f64:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX11-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v3f64:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[6:7], v[6:7], v[6:7]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[8:9], v[8:9], v[8:9]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[2:3]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[10:11], v[10:11], v[10:11]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[4:5]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[6:7]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[8:9]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[10:11]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <3 x double> @llvm.maximumnum.v3f64(<3 x double> %x, <3 x double> %y)
|
|
ret <3 x double> %result
|
|
}
|
|
|
|
define <3 x double> @v_maximumnum_v3f64_nnan(<3 x double> %x, <3 x double> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v3f64_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v3f64_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v3f64_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v3f64_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[6:7]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[8:9]
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[10:11]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v3f64_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[6:7]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[8:9]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[10:11]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <3 x double> @llvm.maximumnum.v3f64(<3 x double> %x, <3 x double> %y)
|
|
ret <3 x double> %result
|
|
}
|
|
|
|
define <4 x double> @v_maximumnum_v4f64(<4 x double> %x, <4 x double> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v4f64:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX8-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX8-NEXT: v_max_f64 v[12:13], v[12:13], v[12:13]
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX8-NEXT: v_max_f64 v[14:15], v[14:15], v[14:15]
|
|
; GFX8-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX8-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v4f64:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX9-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX9-NEXT: v_max_f64 v[12:13], v[12:13], v[12:13]
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX9-NEXT: v_max_f64 v[14:15], v[14:15], v[14:15]
|
|
; GFX9-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX9-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v4f64:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX10-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX10-NEXT: v_max_f64 v[12:13], v[12:13], v[12:13]
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX10-NEXT: v_max_f64 v[14:15], v[14:15], v[14:15]
|
|
; GFX10-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX10-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v4f64:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[8:9], v[8:9], v[8:9]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[0:1]
|
|
; GFX11-NEXT: v_max_f64 v[10:11], v[10:11], v[10:11]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[2:3]
|
|
; GFX11-NEXT: v_max_f64 v[12:13], v[12:13], v[12:13]
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[4:5]
|
|
; GFX11-NEXT: v_max_f64 v[14:15], v[14:15], v[14:15]
|
|
; GFX11-NEXT: v_max_f64 v[6:7], v[6:7], v[6:7]
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4)
|
|
; GFX11-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v4f64:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[8:9], v[8:9], v[8:9]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[0:1]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[10:11], v[10:11], v[10:11]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[2:3]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[12:13], v[12:13], v[12:13]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[4:5]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[14:15], v[14:15], v[14:15]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[6:7], v[6:7], v[6:7]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[8:9]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[10:11]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[12:13]
|
|
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_4)
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[6:7], v[6:7], v[14:15]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call <4 x double> @llvm.maximumnum.v4f64(<4 x double> %x, <4 x double> %y)
|
|
ret <4 x double> %result
|
|
}
|
|
|
|
define <4 x double> @v_maximumnum_v4f64_nnan(<4 x double> %x, <4 x double> %y) {
|
|
; GFX8-LABEL: v_maximumnum_v4f64_nnan:
|
|
; GFX8: ; %bb.0:
|
|
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX8-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX8-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX8-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX8-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-LABEL: v_maximumnum_v4f64_nnan:
|
|
; GFX9: ; %bb.0:
|
|
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX9-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX9-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX9-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX9-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-LABEL: v_maximumnum_v4f64_nnan:
|
|
; GFX10: ; %bb.0:
|
|
; GFX10-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX10-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX10-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX10-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX10-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-LABEL: v_maximumnum_v4f64_nnan:
|
|
; GFX11: ; %bb.0:
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[8:9]
|
|
; GFX11-NEXT: v_max_f64 v[2:3], v[2:3], v[10:11]
|
|
; GFX11-NEXT: v_max_f64 v[4:5], v[4:5], v[12:13]
|
|
; GFX11-NEXT: v_max_f64 v[6:7], v[6:7], v[14:15]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX12-LABEL: v_maximumnum_v4f64_nnan:
|
|
; GFX12: ; %bb.0:
|
|
; GFX12-NEXT: s_wait_loadcnt_dscnt 0x0
|
|
; GFX12-NEXT: s_wait_expcnt 0x0
|
|
; GFX12-NEXT: s_wait_samplecnt 0x0
|
|
; GFX12-NEXT: s_wait_bvhcnt 0x0
|
|
; GFX12-NEXT: s_wait_kmcnt 0x0
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[8:9]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[2:3], v[10:11]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[4:5], v[4:5], v[12:13]
|
|
; GFX12-NEXT: v_max_num_f64_e32 v[6:7], v[6:7], v[14:15]
|
|
; GFX12-NEXT: s_setpc_b64 s[30:31]
|
|
%result = call nnan <4 x double> @llvm.maximumnum.v4f64(<4 x double> %x, <4 x double> %y)
|
|
ret <4 x double> %result
|
|
}
|